GETTING STARTED | CAPABILITIES | KNIME ANALYTICS PLATFORM
My Favorite KNIME Capabilities
It was incredibly hard limiting this to just 5!
There are many amazing things about KNIME, but here are some of my favorites.
KNIME has pretty awesome community members who often share their workflows and components to benefit others. Here you can find a workflow or a component that applies to your specific use case. For instance, a search on the NLP returns the below workflows!
The KNIME Hub is great for idea generation or to help you get unstuck. The KNIME Hub is undoubtedly a nice rabbit hole to explore if you have a few minutes. But heads up, those minutes can quickly turn to hours.
2. Node Pit
In addition to the KNIME Extensions, KNIME users enjoy a myriad of tools installable from the Node Pit. You can get Node Pit by following the “Install via Update Site” instructions found here. The Palladian extension is my favourite Node Pit offering; the ReGex Extractor node has gotten me out of many sticky situations.
You do not have to KNIME alone, thanks to the KNIME Community Forum. If you have a question, there is a high probability someone has already asked that question on the forum, and answers await. The forum was invaluable to me as a new KNIME user, and it certainly helped accelerate my learning. If you cannot find solutions for your specific use case, it is effortless to post a question and quickly benefit from the rich knowledge of the KNIMErs from all around the world.
4. Cross Platform Integration
One of the major appeals of KNIME is that coding is not required. But if you want to/or need to code, KNIME has your back! This flexibility is one of things I love about KNIME. In fact I call KNIME the hippy of software since it just loves to get along with everyone.
For one use case on sentiment analysis, I pull tweets via the twitter connector and twitter API nodes, leverage KNIME text processing nodes to clean my tweets, and then publish my data to Power BI, all within the KNIME platform.
I would like to give a shout out to @Takbb from the KNIME Hub. My Python node was failing, seemingly due to emojis contained in the tweet, but I was able to filter out these emojis from the text leveraging his String Emoji Filter component. KNIME Hub to the rescue!
5. Flow Variables and Loops
Technically, these are two things, but they work exceptionally well as a combo, so I think I can get away with listing them both as a capability 😜.
The combination of flow variables and loops are a masterful symphony of automation.
I genuinely get happy whenever I use these two and reflect on all the time I am saving and all the stress I am avoiding.
The below screenshot is part of a workflow metanode I built to monitor customer retention for a food delivery app. With this combination of loops and flow variables, I was able to see what percentage of customers who ordered in a particular week were retained in the future.
I used a Group Loop to group the customers who ordered each week. I then pass the last date in the data as a flow variable which I use to filter the rest of the data to only include orders in the future. Customer retention was a vital KPI for us, and we were able to calculate and monitor this thanks to KNIME.
Alright, those are my top KNIME capabilities. Next up, I’ll share my top KNIME tips.
